can distance change the sound of a telephone ringtone (telepathy experiment)?

You seriously expect people to watch an HOUR AND FORTY MINUTE video so we can answer some ridiculous question about telepathy? I can't imagine even inmates serving multiple life sentences having enough spare time for such a useless pursuit.

I didn't watch the video, but the usual way to debunk these kinds of "experiments" is to firstly statistically analyse them.

In the experiment you described, you're dealing with probabilities (you have a one in four chance of just randomly guessing who the right caller is without using any magical telepathy or anything). You might think that if the experimental data deviates from the expected probability (for example, the data says that the participants guessed who was calling 2/4 times instead of the expected probability of 1/4 times) then it would be statistically significant, but this isn't what you look for. Instead, you need to see if the data deviates from something called the "null result" http://en.wikipedia.org/wiki/Null_hypothesis

For instance, if I roll a die, I know there is a 1/6 chance of landing on the number "five". But, if I were to roll the die 12 times and I exactly calculated a 1/6 probability of landing on five (meaning each number was rolled exactly twice), then that would be pretty remarkable. If you actually rolled it 12 times you'd probably find that you landed on "5" four out of 12 times or something. This is because your sample size was small, so you wouldn't expect your results to be very close to the expected probability anyway.

So in the experiment you described, the sample size was tiny, so the null result is very large. If you really want to, you can look at the experimental data and see if it deviates from the null result. If it does deviate from the null result, you would need that deviation to be very large in order for it to be statistically significant (read http://en.wikipedia.org/wiki/Significance_level ).

The other thing is to compare the results to repeated experiments. If there is a statistical significance, and that same statistical significance continues to show up in repeated experiments (done by different people), then that experiment might have been legitimate. Otherwise, the guy in the video is just another crank.

A lot of the time, people do these kinds of pseudoscientific experiments and claim that they're legitimate, but it's easily debunked by a simple statistical analysis like this. There's a reason why this kind of stuff isn't done by real scientists.